A method and system for heliostat direction correction for solar thermal power generation

Abstract

The application discloses a heliostat direction correction method and system for photo-thermal power generation, and relates to the technical field of data processing. The method comprises the following steps: acquiring a heliostat array of a target area, and acquiring layout information of the heliostat array; obtaining a plurality of target area grids; respectively calling a plurality of historical correction data of the plurality of target area grids, performing volume division on the plurality of target area grids, and obtaining a first area grid set and a second area grid set; respectively performing image acquisition on the heliostat array of the first area grid set and the second area grid set through an image acquisition device, and obtaining a first image set and a second image set; inputting into a direction correction model, and outputting a first correction parameter set and a second correction parameter set; and performing direction correction on the heliostat array. The application solves the technical problem that the existing heliostat direction correction has low intelligent degree and poor accuracy, and achieves the technical effect of improving correction accuracy.

Description

A method and system for heliostat orientation correction in concentrated solar power generation Technical Field

[0001] This invention relates to the field of data processing technology, and specifically to a method and system for correcting the orientation of heliostats used in solar thermal power generation. Background Technology

[0002] Heliostats are widely used in concentrated solar power (CSP), but in practice, their orientation correction is mainly performed by technicians, which is inefficient. Existing technologies suffer from low levels of intelligence and poor accuracy in heliostat orientation correction. Summary of the Invention

[0003] This application provides a method and system for heliostat orientation correction in solar thermal power generation, which addresses the technical problems of low intelligence and poor accuracy in existing heliostat orientation correction technologies.

[0004] In view of the above problems, this application provides a method and system for heliostat orientation correction for concentrated solar power generation.

[0005] The first aspect of this application provides a heliostat orientation correction method for concentrated solar power generation, the method comprising:

[0006] Acquire the heliostat array in the target area and acquire the layout information of the heliostat array, wherein the layout information includes the distribution location and number of heliostats;

[0007] Based on the deployment information, the target area is divided into regional grids to obtain multiple target area grids;

[0008] Multiple historical correction data of the multiple target region grids are retrieved respectively, and the multiple target region grids are divided into volumetric values ​​to obtain a first region grid set and a second region grid set, wherein the volume of the first region grid set is greater than the volume of the second region grid set.

[0009] The image acquisition device acquires images of the heliostat arrays of the first and second region grid sets respectively, and obtains the first image set and the second image set.

[0010] The first image set and the second image set are respectively input into the orientation correction model, and the first correction parameter set and the second correction parameter set are output.

[0011] The orientation of the heliostat array is corrected based on the first set of correction parameters and the second set of correction parameters.

[0012] A second aspect of this application provides a heliostat orientation correction system for concentrated solar power generation, the system comprising:

[0013] The deployment information acquisition module is used to acquire the heliostat array in the target area and acquire the deployment information of the heliostat array, wherein the deployment information includes the distribution location of the heliostats and the number of heliostats.

[0014] A region grid acquisition module is used to divide the target area into region grids according to the deployment information, thereby obtaining multiple target region grids.

[0015] A volume division module is used to retrieve multiple historical correction data of the multiple target region grids respectively, divide the multiple target region grids into volume divisions, and obtain a first region grid set and a second region grid set, wherein the volume of the first region grid set is greater than the volume of the second region grid set.

[0016] An image acquisition module is used to acquire images of the heliostat arrays of the first region grid set and the second region grid set respectively through an image acquisition device, thereby obtaining a first image set and a second image set.

[0017] The correction parameter output module is used to input the first image set and the second image set into the orientation correction model respectively, and output the first correction parameter set and the second correction parameter set;

[0018] An orientation correction module is used to perform orientation correction on the heliostat array according to the first set of correction parameters and the second set of correction parameters.

[0019] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0020] This application acquires information about the heliostat array in a target area, including its layout information such as the location and number of heliostats. Based on this information, the target area is divided into multiple target area grids. Historical correction data for each of these grids is retrieved, and the grids are further subdivided into a first grid set and a second grid set, where the first grid set is larger than the second grid set. Images of the heliostat arrays within both the first and second grid sets are then acquired using an image acquisition device, resulting in a first image set and a second image set. These images are then input into a direction correction model, which outputs a first set of correction parameters and a second set of correction parameters. The direction of the heliostat array is then corrected based on these parameters. This achieves the technical effect of improving the intelligence level of heliostat direction correction. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 is a schematic flowchart of a heliostat orientation correction method for solar thermal power generation provided in an embodiment of this application;

[0023] Figure 2 is a schematic diagram of a heliostat orientation correction system for solar thermal power generation provided in an embodiment of this application.

[0024] Figure labeling: Layout information acquisition module 11, area grid acquisition module 12, volume division module 13, image acquisition module 14, correction parameter output module 15, orientation correction module 16. Detailed Implementation

[0025] This application provides a method and system for heliostat orientation correction in solar thermal power generation, which addresses the technical problems of low intelligence and poor accuracy in existing heliostat orientation correction technologies.

[0026] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0027] It should be noted that the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such process, method, product, or device.

[0028] Example 1

[0029] As shown in Figure 1, this application provides a heliostat orientation correction method for concentrated solar power generation, wherein the method includes:

[0030] Step S100: Obtain the heliostat array in the target area and obtain the layout information of the heliostat array, wherein the layout information includes the distribution location of the heliostats and the number of heliostats.

[0031] In one possible embodiment, the target area is any area where heliostats are deployed. The heliostat array is a plurality of heliostats arranged in an orderly manner within the target area according to a preset arrangement method. The deployment information describes the arrangement of the heliostat array within the target area, including the location and number of heliostats. The location describes the position of the heliostats within the target area, and the number of heliostats is the total number of heliostats present in the target area.

[0032] Step S200: Divide the target area into regional grids according to the deployment information to obtain multiple target area grids;

[0033] In one possible embodiment, the target area is divided into grids based on the location and number of heliostats in the deployment information. Preferably, the target area is first divided into equal grids to obtain a primary grid result. Then, based on the number of heliostats in the primary grid result, the primary grid results with a number lower than a preset distribution are merged to obtain multiple target area grids. These multiple target area grids rationally divide the target area, thereby improving the accuracy of subsequent analysis.

[0034] Step S300: Retrieve multiple historical correction data of the multiple target area grids respectively, divide the multiple target area grids into volumetric values, and obtain a first area grid set and a second area grid set, wherein the volume of the first area grid set is greater than the volume of the second area grid set;

[0035] Furthermore, step S300 in this embodiment of the application also includes:

[0036] Step S310: Determine the number of heliostats in the multiple target area grids based on the layout information of the heliostat array;

[0037] Step S320: Use the ratio of the number of heliostats in each target region grid to the number of heliostats in the multiple target region grids as the first volume coefficient to generate multiple first volume coefficients;

[0038] Step S330: Generate multiple second volume coefficients based on the deviation values ​​between the multiple historical correction data and the preset correction data threshold;

[0039] Step S340: Divide the multiple target region grids into multiple first volume coefficients and multiple second volume coefficients to obtain a first region grid set and a second region grid set.

[0040] Furthermore, step S300 in this embodiment of the application also includes:

[0041] Step S350: Obtain the preset weight allocation values ​​based on the expert survey method;

[0042] Step S360: Based on multiple first volume coefficients and multiple second volume coefficients, perform weighted calculations according to preset volume weight allocation values ​​to obtain multiple volume coefficients;

[0043] Step S370: Obtain the average value of the coefficients based on multiple volume coefficients, iterate through the multiple volume coefficients and compare them with the average value, classify the target area grids with volume coefficients lower than the average value into the second area grid set, and classify the target area grids with volume coefficients higher than the average value into the first area grid set.

[0044] In the embodiments of this application, multiple historical correction data of multiple target area grids are acquired, and the multiple target area grids are divided into sizes according to the information contained in the data to obtain a first area grid set and a second area grid set, wherein the size of the first area grid set is larger than the size of the second area grid set.

[0045] In one possible embodiment, the number of heliostats contained within the plurality of target area grids is determined based on the deployment information of the heliostat array, thereby obtaining the quantity information of the plurality of heliostats. Then, the ratio of the quantity information of heliostats in each target area grid to the total quantity information of the plurality of heliostats is used as a first volume coefficient to generate a plurality of first volume coefficients. Based on the magnitude of the deviation between the plurality of historical correction data and a preset correction data threshold, a plurality of second volume coefficients are generated.

[0046] In one possible embodiment, a preset volume weight allocation value is obtained by using an expert survey method, that is, selecting any number of experts to set volume weight allocation values, and then averaging the results of multiple expert settings. Then, based on multiple first volume coefficients and multiple second volume coefficients, a weighted calculation is performed according to the preset volume weight allocation value to obtain multiple volume coefficients. These volume coefficients reflect the amount of data that needs to be processed during correction in different regional grid sets.

[0047] In the embodiments of this application, the average value of the coefficients is obtained based on multiple volume coefficients, and the multiple volume coefficients are compared with the average value of the coefficients respectively. The target area grid with the volume coefficient lower than the average value of the coefficients is classified into the second area grid set, and the target area grid with the volume coefficient higher than the average value of the coefficients is classified into the first area grid set.

[0048] Step S400: The image acquisition device acquires images of the heliostat arrays of the first region grid set and the second region grid set respectively to obtain the first image set and the second image set;

[0049] In one possible embodiment, the image acquisition device acquires images reflected in the heliostat array. A first image set reflects the deviation of the sun's spot from the center of the heliostat within a first grid region. A second image set reflects the deviation of the sun's spot from the center of the heliostat within a second grid region. Obtaining the first and second image sets lays the groundwork for subsequent image correction analysis.

[0050] Step S500: Input the first image set and the second image set into the orientation correction model respectively, and output the first correction parameter set and the second correction parameter set;

[0051] Step S600: Perform orientation correction on the heliostat array according to the first set of correction parameters and the second set of correction parameters.

[0052] Furthermore, step S500 in this embodiment of the application also includes:

[0053] Step S510: Extract from the first image set according to the first preset extraction frequency to obtain the first corrected image set;

[0054] Step S520: Extract from the second image set according to the second preset extraction frequency to obtain the second corrected image set;

[0055] Step 530: Input the first set of corrected images into the fast processing channel of the orientation correction model to obtain the first set of correction parameters;

[0056] Step S540: Input the second set of corrected images into the slow processing channel of the orientation correction model to obtain the second set of correction parameters.

[0057] Furthermore, step S500 in this embodiment of the application also includes:

[0058] Step S550: The orientation correction model includes a fast processing channel and a slow processing channel;

[0059] Step S560: Obtain a set of multiple sample first-corrected images and a set of multiple sample first-corrected parameters as the first construction data to generate the fast processing channel;

[0060] Step S570: Obtain a set of multiple sample second-corrected images and a set of multiple sample second-corrected parameters as second construction data to generate the slow processing channel.

[0061] Furthermore, step S560 in this embodiment of the application also includes:

[0062] Step S561: Divide the first constructed data according to a preset division ratio to obtain a training set and a validation set;

[0063] Step S562: Train the framework built on the BP neural network using the training set until it converges;

[0064] Step S563: Validate the converged fast processing channel using the validation set. If the validation passes, the fast processing channel is obtained.

[0065] Furthermore, step S570 in this embodiment of the application also includes:

[0066] Step S571: Construct a correction mapping relationship based on the multiple sets of second-corrected sample images and the multiple sets of second-corrected sample parameters;

[0067] Step S572: Generate the slow processing channel according to the correction mapping relationship.

[0068] In one possible embodiment, the orientation correction model is a functional model for intelligently analyzing the correction parameters of heliostats based on image set analysis, including a fast processing channel and a slow processing channel. The fast processing channel processes a first set of correction images, analyzing the heliostat correction parameters corresponding to a large target region grid. The slow processing channel processes a second set of correction images, analyzing the heliostat correction parameters corresponding to a smaller target region grid. The first set of correction parameters is used to perform orientation correction on the heliostats within the first region grid set. The second set of correction parameters is used to perform orientation correction on the heliostats within the second region grid set.

[0069] In one possible embodiment, the first preset extraction frequency is the number of image frames extracted from the first image set per unit time, and the second preset extraction frequency is the number of image frames extracted from the second image set per unit time. The first preset extraction frequency is less than the second preset extraction frequency. That is, the number of image frames extracted from the first image set per unit time is less than the number of image frames extracted from the second image set per unit time. Therefore, by extracting from the first image set according to the first preset extraction frequency, a first corrected image set is obtained; and by extracting from the second image set according to the second preset extraction frequency, a second corrected image set is obtained. This improves the efficiency of the correction analysis. The first region grid set is large in size and requires more correction; therefore, the image changes between adjacent frames are large during correction image analysis, and more images are not needed to ensure accurate correction. The second region grid set is smaller in size and requires higher correction accuracy; therefore, more images are needed for analysis. By inputting the first corrected image set into the fast processing channel of the orientation correction model, a first correction parameter set is obtained; and by inputting the second corrected image set into the slow processing channel of the orientation correction model, a second correction parameter set is obtained.

[0070] In one possible embodiment, multiple sets of sample first-corrected images and multiple sets of sample first-corrected parameters are acquired as first-construction data to generate the fast processing channel. The first-construction data is divided according to a preset division ratio to obtain a training set and a validation set. The framework built on a BP neural network is trained using the training set until convergence. The multiple sets of sample first-corrected images from the validation set are input into the fast processing channel. The obtained multiple sets of sample validation first-corrected parameters are compared with the multiple sets of sample first-corrected parameters. The ratio of the number of successfully matched parameter sets to the number of multiple sets of sample first-corrected parameters is used as the validation accuracy. If the validation accuracy meets the preset validation accuracy, the validation is passed, and the fast processing channel is obtained. A calibration mapping relationship is constructed based on the multiple sets of sample second-corrected images and multiple sets of sample second-corrected parameters. This calibration mapping relationship is used as the construction logic for the slow processing channel, thereby obtaining the slow processing channel. The orientation of the heliostat array is corrected according to the first and second sets of calibration parameters.

[0071] In summary, the embodiments of this application have at least the following technical effects:

[0072] This application acquires information about the heliostat array in a target area, divides the target area into regional grids based on the deployment information, retrieves multiple historical correction data from these target regional grids, performs volumetric subdivision, and then uses an image acquisition device to acquire images of the heliostat arrays in a first and second regional grid set, obtaining a first image set and a second image set. Then, using an intelligent orientation correction model, a first correction parameter set and a second correction parameter set are obtained. Finally, the orientation of the heliostat array is corrected based on the first and second correction parameter sets. This achieves the technical effect of improving the intelligence level of orientation correction and enhancing the correction quality.

[0073] Example 2

[0074] Based on the same inventive concept as the heliostat orientation correction method for concentrated solar power generation described in the foregoing embodiments, as shown in Figure 2, this application provides a heliostat orientation correction system for concentrated solar power generation. The system and method embodiments in this application are based on the same inventive concept. The system includes:

[0075] The deployment information acquisition module 11 is used to acquire the heliostat array in the target area and acquire the deployment information of the heliostat array, wherein the deployment information includes the distribution location of the heliostats and the number of heliostats.

[0076] The region grid acquisition module 12 is used to divide the target area into region grids according to the deployment information to obtain multiple target region grids.

[0077] The volume division module 13 is used to retrieve multiple historical correction data of the multiple target area grids respectively, divide the multiple target area grids into volumes, and obtain a first area grid set and a second area grid set, wherein the volume of the first area grid set is greater than the volume of the second area grid set.

[0078] Image acquisition module 14 is used to acquire images of the heliostat array of the first region grid set and the second region grid set respectively through the image acquisition device to obtain the first image set and the second image set;

[0079] The correction parameter output module 15 is used to input the first image set and the second image set into the orientation correction model respectively, and output the first correction parameter set and the second correction parameter set;

[0080] The orientation correction module 16 is used to perform orientation correction on the heliostat array according to the first set of correction parameters and the second set of correction parameters.

[0081] Furthermore, the volume division module 13 is used to perform the following method:

[0082] The number of heliostats in the multiple target area grids is determined based on the deployment information of the heliostat array.

[0083] The ratio of the number of heliostats in each target region grid to the total number of heliostats in the multiple target region grids is used as the first volume coefficient to generate multiple first volume coefficients.

[0084] Based on the magnitude of the deviation between the multiple historical correction data and the preset correction data threshold, multiple second volume coefficients are generated;

[0085] Based on multiple first volume coefficients and multiple second volume coefficients, the multiple target region grids are divided into volumetric values ​​to obtain a first region grid set and a second region grid set.

[0086] Furthermore, the volume division module 13 is used to perform the following method:

[0087] The preset weight allocation values ​​were obtained based on expert surveys.

[0088] Based on multiple first volume coefficients and multiple second volume coefficients, multiple volume coefficients are obtained by weighted calculation according to preset volume weight allocation values.

[0089] The average value of the coefficients is obtained based on multiple volume coefficients. The multiple volume coefficients are iterated and compared with the average value. Target area grids with volume coefficients lower than the average value are classified into the second area grid set, and target area grids with volume coefficients higher than the average value are classified into the first area grid set.

[0090] Furthermore, the correction parameter output module 15 is used to perform the following method:

[0091] The first image set is extracted according to the first preset extraction frequency to obtain the first corrected image set;

[0092] The second image set is extracted according to the second preset extraction frequency to obtain the second corrected image set;

[0093] The first set of corrected images is input into the fast processing channel of the orientation correction model to obtain the first set of correction parameters;

[0094] The second set of corrected images is input into the slow processing channel of the orientation correction model to obtain the second set of correction parameters.

[0095] Furthermore, the correction parameter output module 15 is used to perform the following method:

[0096] The orientation correction model includes a fast processing channel and a slow processing channel;

[0097] The fast processing channel is generated by acquiring a set of first-corrected images of multiple samples and a set of first-corrected parameters of multiple samples as the first construction data.

[0098] Multiple sets of second-corrected images and multiple sets of second-corrected parameters for samples are acquired as second-construction data to generate the slow-speed processing channel.

[0099] Furthermore, the correction parameter output module 15 is used to perform the following method:

[0100] The first constructed data is divided according to a preset division ratio to obtain a training set and a validation set;

[0101] The framework built on the BP neural network is trained using the training set until it converges.

[0102] The converged fast processing channel is verified using a validation set. If the verification passes, the fast processing channel is obtained.

[0103] Furthermore, the correction parameter output module 15 is used to perform the following method:

[0104] A correction mapping relationship is constructed based on the multiple sets of second-corrected sample images and the multiple sets of second-corrected sample parameters;

[0105] The slow processing channel is generated based on the correction mapping relationship.

[0106] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0107] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0108] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for heliostat orientation correction in concentrated solar power generation, characterized in that, The method includes: acquiring the heliostat array of the target area and acquiring the layout information of the heliostat array, wherein the layout information includes the distribution location and number of heliostats; dividing the target area into regional grids according to the layout information to obtain multiple target regional grids; retrieving multiple historical correction data of the multiple target regional grids respectively, and dividing the multiple target regional grids into volumetric sizes to obtain a first regional grid set and a second regional grid set, wherein the volume of the first regional grid set is larger than the volume of the second regional grid set; and acquiring images of the heliostat arrays of the first regional grid set and the second regional grid set respectively using an image acquisition device to obtain a first image. The method comprises: extracting a first image set and a second image set; inputting the first image set and the second image set into an orientation correction model, respectively, and outputting a first correction parameter set and a second correction parameter set; and performing orientation correction on the heliostat array according to the first correction parameter set and the second correction parameter set; the method includes: extracting the first image set at a first preset extraction frequency to obtain a first corrected image set; extracting the second image set at a second preset extraction frequency to obtain a second corrected image set; inputting the first corrected image set into a fast processing channel of the orientation correction model to obtain a first correction parameter set; and inputting the second corrected image set into a slow processing channel of the orientation correction model to obtain a second correction parameter set.

2. The method as described in claim 1, characterized in that, The method includes: determining the number of heliostats in the plurality of target area grids based on the deployment information of the heliostat array; using the ratio of the number of heliostats in each target area grid to the total number of heliostats as a first volume coefficient to generate a plurality of first volume coefficients; generating a plurality of second volume coefficients based on the deviation values ​​between the plurality of historical correction data and a preset correction data threshold; and dividing the plurality of target area grids into volume fractions based on the plurality of first volume coefficients and the plurality of second volume coefficients to obtain a first area grid set and a second area grid set.

3. The method as described in claim 2, characterized in that, The method includes: obtaining a preset volume weight allocation value based on an expert survey; performing a weighted calculation based on multiple first volume coefficients and multiple second volume coefficients according to the preset volume weight allocation value to obtain multiple volume coefficients; obtaining the average value of the multiple volume coefficients; comparing the multiple volume coefficients with the average value; classifying target area grids with volume coefficients lower than the average value into a second area grid set; and classifying target area grids with volume coefficients higher than the average value into a first area grid set.

4. The method as described in claim 1, characterized in that, The method includes: the orientation correction model includes a fast processing channel and a slow processing channel; acquiring a set of multiple sample first correction images and a set of multiple sample first correction parameters as first construction data to generate the fast processing channel; acquiring a set of multiple sample second correction images and a set of multiple sample second correction parameters as second construction data to generate the slow processing channel.

5. The method as described in claim 4, characterized in that, The method includes: dividing the first constructed data according to a preset division ratio to obtain a training set and a validation set; using the training set to train the framework built on the BP neural network until it converges; and using the validation set to validate the converged fast processing channel. If the validation passes, the fast processing channel is obtained.

6. The method as described in claim 4, characterized in that, The method includes: constructing a correction mapping relationship based on the multiple sample second correction image set and the multiple sample second correction parameter set; and generating the slow processing channel according to the correction mapping relationship.

7. A heliostat orientation correction system for concentrated solar power generation, characterized in that, The system includes: a deployment information acquisition module, used to acquire the heliostat array of the target area and the deployment information of the heliostat array, wherein the deployment information includes the distribution location and number of heliostats; a region grid acquisition module, used to divide the target area into region grids according to the deployment information to obtain multiple target region grids; a volume division module, used to retrieve multiple historical correction data of the multiple target region grids respectively, and divide the multiple target region grids into volume divisions to obtain a first region grid set and a second region grid set, wherein the volume of the first region grid set is greater than the volume of the second region grid set; and an image acquisition module, used to capture images of the first region grid set and the second region grid set respectively through an image acquisition device. The system acquires images from a heliostat array of lattice sets to obtain a first image set and a second image set. A correction parameter output module is used to input the first and second image sets into a direction correction model, respectively, and output a first correction parameter set and a second correction parameter set. A direction correction module is used to perform direction correction on the heliostat array based on the first and second correction parameter sets. The system extracts images from the first image set at a first preset extraction frequency to obtain a first corrected image set; extracts images from the second image set at a second preset extraction frequency to obtain a second corrected image set; inputs the first corrected image set into the fast processing channel of the direction correction model to obtain a first correction parameter set; and inputs the second corrected image set into the slow processing channel of the direction correction model to obtain a second correction parameter set.

Images